From 81ec9d3d031c027d32a31b08c881064fc28f78cb Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 2 Sep 2026 12:36:54 +0000 Subject: [PATCH] skills(automation): delete the planned-eval stub and drop its ceiling row MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit #14296 item 2 = A (comment 5507419465): the planned-eval stubs are deleted and their ceilings step down; real fixtures stay. This is round 2 of #14308 — round 1 (PR #14467, merged) implemented every finding in skills/objectstack-automation/SKILL.md and deferred AUTO-B-04, the evals/README.md planned-structure stub, to this ruling. Deletes skills/objectstack-automation/evals/README.md (414 tokens — the template listing planned eval files that do not exist) and drops its one ceiling row from scripts/check-skills-token-ratchet.mjs. The real fixture evals/approvals/test-revise-loop.md and its ratchet row are untouched. Co-Authored-By: Claude Fable 5.1 Claude-Session: https://claude.ai/code/session_01LraLgQVGq8egUwfYZpbYt1 --- scripts/check-skills-token-ratchet.mjs | 1 - skills/objectstack-automation/evals/README.md | 43 ------------------- 2 files changed, 44 deletions(-) delete mode 100644 skills/objectstack-automation/evals/README.md diff --git a/scripts/check-skills-token-ratchet.mjs b/scripts/check-skills-token-ratchet.mjs index db0e70145b..d4aad2565d 100644 --- a/scripts/check-skills-token-ratchet.mjs +++ b/scripts/check-skills-token-ratchet.mjs @@ -431,7 +431,6 @@ export const CEILINGS = new Map([ ['skills/objectstack-query/rules/pagination.md', 1382], // objectstack-automation - ['skills/objectstack-automation/evals/README.md', 414], ['skills/objectstack-automation/evals/approvals/test-revise-loop.md', 1329], // objectstack-ui — the two authored eval files; its `references/react-blocks.md` diff --git a/skills/objectstack-automation/evals/README.md b/skills/objectstack-automation/evals/README.md deleted file mode 100644 index 6a0f6880e0..0000000000 --- a/skills/objectstack-automation/evals/README.md +++ /dev/null @@ -1,43 +0,0 @@ -# Evaluation Tests (evals/) - -Evaluation tests (evals) validate that AI assistants correctly understand and -apply the rules defined in this skill when generating automation metadata — -flows, approval chains, triggers, and scheduled sweeps. - -## Current evals - -| Eval | Covers | -|:-----|:-------| -| [approvals/test-revise-loop.md](./approvals/test-revise-loop.md) | ADR-0044 send-back-for-revision: the `revise` branch, the `approval_revise` window, and the resubmit edge `type: 'back'` | - -## Planned structure - -Future evals extend the same layout, one folder per automation concern: - -``` -evals/ -├── approvals/ -│ ├── test-revise-loop.md ← implemented -│ └── test-quorum-behaviors.md (planned — quorum / per_group sign-off) -├── flows/ -│ ├── test-schedule-binding.md (planned — cadence on the start node, never top-level) -│ └── test-decision-edges.md (planned — edge-condition branching; unguarded edges run in parallel) -└── triggers/ - └── test-time-relative.md (planned — timeRelative sweep vs record-change date-equality) -``` - -## Format - -Each eval file contains: -1. **Scenario** — Description of the task -2. **Expected Output** — Correct implementation -3. **Common Mistakes** — Incorrect patterns to avoid -4. **Validation Criteria** — How to score the output - -## Contributing - -When adding evals: -1. Each eval should test a single, specific rule or pattern from `SKILL.md` -2. Include both positive (correct) and negative (incorrect) examples -3. Name the `SKILL.md` section the eval enforces -4. Use realistic scenarios from actual ObjectStack projects